def initDatabase(dbFilename, iFilename, clobber=False):
db = Alchemy()
gene_table = Table("Gene", db.metadata,
Column('id', Integer, primary_key=True),
Column('geneId', Text),
Column('chrom', Text, index=True),
Column('start', Integer, index=True),
Column('end', Integer, index=True),
Column('strand', String(1))
)
mapper(Gene, gene_table)
populateDb = clobber or not os.path.exists(dbFilename)
session = db.startSession('sqlite:///%s' % dbFilename)
if populateDb:
iFile = open(iFilename)
headers = iFile.readline()
strandTable = {'1': '+', '-1': '-'}
for i,line in enumerate(iFile):
if (i % 1000): progressMessage('# genes %s', i)
tokens = line.strip().split('\t')
g = Gene(None, tokens[0], tokens[5], int(tokens[6]), int(tokens[7]),
strandTable[tokens[8]])
session.save(g)
progressMessage('# genes %s', i)
session.commit()
return session
def initDatabase(dbFilename, iFilename, clobber=False):
db = Alchemy()
gene_table = Table(
"Gene",
db.metadata,
Column("id", Integer, primary_key=True),
Column("geneId", Text),
Column("chrom", Text, index=True),
Column("start", Integer, index=True),
Column("end", Integer, index=True),
Column("strand", String(1)),
)
mapper(Gene, gene_table)
populateDb = clobber or not os.path.exists(dbFilename)
session = db.startSession("sqlite:///%s" % dbFilename)
if populateDb:
iFile = open(iFilename)
headers = iFile.readline()
strandTable = {"1": "+", "-1": "-"}
for i, line in enumerate(iFile):
if i % 1000:
progressMessage("# genes %s", i)
tokens = line.strip().split("\t")
g = Gene(None, tokens[0], tokens[5], int(tokens[6]), int(tokens[7]), strandTable[tokens[8]])
session.save(g)
progressMessage("# genes %s", i)
session.commit()
return session
def initDatabase(dbFilename, iFilename, clobber=False):
db = Alchemy()
gene_table = Table("Gene", db.metadata,
Column('id', Integer, primary_key=True),
Column('geneId', Text),
Column('chrom', Text, index=True),
Column('start', Integer, index=True),
Column('end', Integer, index=True),
Column('strand', String(1))
)
mapper(Gene, gene_table)
populateDb = clobber or not os.path.exists(dbFilename)
session = db.startSession('sqlite:///%s' % dbFilename)
if populateDb:
iFile = open(iFilename)
headers = iFile.readline()
strandTable = {'1': '+', '-1': '-'}
for i,line in enumerate(iFile):
if (i % 1000): progressMessage('# genes %s', i)
tokens = line.strip().split('\t')
g = Gene(None, tokens[0], tokens[5], int(tokens[6]), int(tokens[7]),
strandTable[tokens[8]])
session.save(g)
progressMessage('# genes %s', i)
session.commit()
return session
def build(self, clobber=False, separator='!'):
"""Build index file.
@keyword clobber: Overwrite existing index file (default=False)
"""
if clobber:
try:
self.cursor.execute('drop table Offsets;')
except:
pass
if not self.isIndexed():
schema = """
CREATE TABLE Offsets (
id INTEGER,
accession TEXT,
offset INTEGER
);
create index idx_offsets_id on Offsets (id);
create index idx_offsets_accession on Offsets (accession);
"""
self.connection.executescript(schema)
tmpFile = tempfile.NamedTemporaryFile()
for i,(accession,offset) in enumerate(self._byteOffsetGenerator()):
print >> tmpFile, "%i%s%s%s%i" % (i,separator,accession,separator,offset)
if i % 1000==0:
progressMessage("Sequences: %s", i+1)
progressMessage("Sequences: %s\n", i+1)
tmpFile.flush()
cmd = """sqlite3 -separator '%s' %s '.import "%s" Offsets'""" \
% (separator, self.idxFilename, tmpFile.name)
os.system(cmd)
tmpFile.close()
def build(self, clobber=False, separator='!'):
"""Build index file.
@keyword clobber: Overwrite existing index file (default=False)
"""
if clobber:
try:
self.cursor.execute('drop table Offsets;')
except:
pass
if not self.isIndexed():
schema = """
CREATE TABLE Offsets (
id INTEGER,
accession TEXT,
offset INTEGER
);
create index idx_offsets_id on Offsets (id);
create index idx_offsets_accession on Offsets (accession);
"""
self.connection.executescript(schema)
tmpFile = tempfile.NamedTemporaryFile()
for i, (accession,
offset) in enumerate(self._byteOffsetGenerator()):
print >> tmpFile, "%i%s%s%s%i" % (i, separator, accession,
separator, offset)
if i % 1000 == 0:
progressMessage("Sequences: %s", i + 1)
progressMessage("Sequences: %s\n", i + 1)
tmpFile.flush()
cmd = """sqlite3 -separator '%s' %s '.import "%s" Offsets'""" \
% (separator, self.idxFilename, tmpFile.name)
os.system(cmd)
tmpFile.close()
# contigData[chrom].add_interval(contig)
# except KeyError:
# contigData[chrom] = Intersecter()
# contigData[chrom].add_interval(contig)
print 'Parse genes'
iFilename = '/Users/papenfuss/databases/platypus/ensembl/Release50/mart_names_locations.txt'
iFile = open(iFilename)
headers = iFile.readline()
annotated = set()
for i,line in enumerate(iFile):
if (i % 1000)==0:
progressMessage('# genes %s', i)
tokens = line.strip().split('\t')
geneId = tokens[0]
transId = tokens[1]
name = tokens[3]
chrom = tokens[5]
start = int(tokens[6])
end = int(tokens[7])
strand = {'1': '+', '-1': '-'}[tokens[8]]
try:
for contig in contigData[chrom].find(start-500, end+500):
annotated.add(contig.value[0])
except:
pass
iFilename = os.path.join(iDir, 'mart_names_locations.txt')
dbFilename = os.path.join(iDir, 'mart_names_locations.sqlite')
# dbFilename = ':memory:'
session = initDatabase(dbFilename, iFilename)
# 2a. Parse read alignment file and has results
# 2b. Write out unannotated reads
flankSize = 1000
maqFilename = '/Users/papenfuss/databases/platypus/venom/solexa/mapview.txt'
data = {}
unannFile = open('unannotated.txt', 'w')
multFile = open('multiple.txt', 'w')
multFile2 = open('multiple_gene.txt', 'w')
for i,m in enumerate(MaqViewFile(maqFilename)):
if (i % 1000)==0: progressMessage('# maq alns %s', i)
q = session.query(Gene).filter(Gene.chrom==m.chrom) \
.filter(Gene.start<m.start+flankSize) \
.filter(Gene.end>m.start+32-flankSize).all()
if len(q)==0:
print >> unannFile, m
continue
elif len(q)>1:
x = set([r.geneId for r in q])
if len(x)>1:
print >> multFile, m
print >> multFile2, "%s\t%s" % (m.name, ','.join(x))
continue
try:
dbFilename = os.path.join(iDir, "mart_names_locations.sqlite")
# dbFilename = ':memory:'
session = initDatabase(dbFilename, iFilename)
# 2a. Parse read alignment file and has results
# 2b. Write out unannotated reads
flankSize = 1000
maqFilename = "/Users/papenfuss/databases/platypus/venom/solexa/mapview.txt"
data = {}
unannFile = open("unannotated.txt", "w")
multFile = open("multiple.txt", "w")
multFile2 = open("multiple_gene.txt", "w")
for i, m in enumerate(MaqViewFile(maqFilename)):
if (i % 1000) == 0:
progressMessage("# maq alns %s", i)
q = (
session.query(Gene)
.filter(Gene.chrom == m.chrom)
.filter(Gene.start < m.start + flankSize)
.filter(Gene.end > m.start + 32 - flankSize)
.all()
)
if len(q) == 0:
print >> unannFile, m
continue
elif len(q) > 1:
x = set([r.geneId for r in q])
if len(x) > 1:
print >> multFile, m
hsps_table = createTable(tableName, metadata, h.attributes, h.converters,
indexedAttributes=['subjectId', 'sStart', 'sEnd'])
mapper(HSP, hsps_table)
# Start a session & initialize database
session = createSession(dsn, metadata)
if case in [1,2]:
# Devil 454 reads
for i,line in enumerate(open(iFilename)):
tokens = line.strip().split('\t')
h = HSP(tokens[0:-2])
h.convertBlockToGenomeCoords()
session.save(h)
if (i % 5000)==0:
progressMessage("# HSPs %s", i, n)
session.commit()
progressMessage("# HSPs %s\n", i, n)
session.commit()
elif case==3:
# Platypus 454 reads
for i,h in enumerate(BlastFile(iFilename)):
h.subjectId = h.subjectId.split('|')[1]
h.convertBlockToGenomeCoords()
session.save(h)
if (i % 5000)==0:
progressMessage("# HSPs %s", i, n)
session.commit()
progressMessage("# HSPs %s\n", i, n)
session.commit()
sys.exit(__doc__)
iFilename = '/Users/papenfuss/databases/platypus/venom/solexa/mapview_filtered.txt' # sys.argv[1]
oFilename = '/Users/papenfuss/platy/venom/gbrowse/solexa250.gff' # sys.argv[2]
windowSize = 250
iFile = open(iFilename)
headers = iFile.readline().strip().split('\t')
oFile = open(oFilename, 'w')
chrom = None
lastChrom = None
countDict = {}
for i, line in enumerate(iFile):
if (i % 1000) == 0:
progressMessage('# reads %s', i, 28000000)
tokens = line.strip().split('\t')
d = dict(zip(headers, tokens))
chrom = "%s" % d['chrom']
start = int(d['start'])
# print chrom, start, lastChrom, (chrom!=lastChrom and len(countDict)!=0)
if chrom != lastChrom and len(countDict) != 0:
countData = countDict.items()
countData.sort(key=lambda x: x[0])
for (_chrom, _wStart), _counts in countData:
g = Feature(reference=_chrom,
source='solexa250',
type='tlevel',
start=_wStart,
end=_wStart + windowSize - 1,
#!/usr/bin/env python
"""
loadSolexa.py
Author: Tony Papenfuss
Date: Tue Jun 24 14:27:34 EST 2008
"""
import os, sys
from maq import *
from useful import progressMessage
oFilename = 'tmp/PlatySolexa.txt'
if not os.path.exists(oFilename):
oFile = open(oFilename, 'w')
dataDir = '/Users/papenfuss/databases/platypus/venom/solexa/'
for i, read in enumerate(
MaqViewFile(os.path.join(dataDir, 'mapview.txt'), mQ_cutoff=40)):
if (i % 1000) == 0:
progressMessage("# maq %s", i, 28395347)
tokens = str(read).split('\t')
tokens.append(i)
print >> oFile, "|".join([str(x) for x in tokens])
oFile.close()
progressMessage("# maq %s\n", i, 28395347)
os.system(
"""sqlite3 alignedReads.db '.import "tmp/PlatySolexa.txt" PlatySolexa'""")
import os, sys
from useful import progressMessage
iFilename = sys.argv[1]
oFilename = sys.argv[2]
mQ_cutoff = 40
nSeqs = 280000000
oFile = open(oFilename, 'w')
headers = ['name','chrom','start','strand','mQ','numTied','score','numZeroMismatches']
format = '\t'.join(['%s','%s','%i','%s','%i','%i','%i','%i'])
print >> oFile, '\t'.join(headers)
for i,line in enumerate(open(iFilename)):
tokens = line.strip().split('\t')
mQ = int(tokens[7])
if mQ>=mQ_cutoff:
name = tokens[0]
chrom = tokens[1]
start = int(tokens[2])
strand = tokens[3]
numTied = int(tokens[10])
score = int(tokens[11])
numZeroMismatches = int(tokens[12])
print >> oFile, format % (name,chrom,start,strand,mQ,numTied,score,numZeroMismatches)
if (i % 1000)==0: progressMessage('# maq hits %s', i, nSeqs)
oFile.close()
h.converters,
indexedAttributes=['subjectId', 'sStart', 'sEnd'])
mapper(HSP, hsps_table)
# Start a session & initialize database
session = createSession(dsn, metadata)
if case in [1, 2]:
# Devil 454 reads
for i, line in enumerate(open(iFilename)):
tokens = line.strip().split('\t')
h = HSP(tokens[0:-2])
h.convertBlockToGenomeCoords()
session.save(h)
if (i % 5000) == 0:
progressMessage("# HSPs %s", i, n)
session.commit()
progressMessage("# HSPs %s\n", i, n)
session.commit()
elif case == 3:
# Platypus 454 reads
for i, h in enumerate(BlastFile(iFilename)):
h.subjectId = h.subjectId.split('|')[1]
h.convertBlockToGenomeCoords()
session.save(h)
if (i % 5000) == 0:
progressMessage("# HSPs %s", i, n)
session.commit()
progressMessage("# HSPs %s\n", i, n)
session.commit()
"""
loadSolexa.py
Author: Tony Papenfuss
Date: Tue Jun 24 14:27:34 EST 2008
"""
import os, sys
from maq import *
from useful import progressMessage
oFilename = 'tmp/PlatySolexa.txt'
if not os.path.exists(oFilename):
oFile = open(oFilename, 'w')
dataDir = '/Users/papenfuss/databases/platypus/venom/solexa/'
for i,read in enumerate(MaqViewFile(os.path.join(dataDir, 'mapview.txt'), mQ_cutoff=40)):
if (i % 1000)==0:
progressMessage("# maq %s", i, 28395347)
tokens = str(read).split('\t')
tokens.append(i)
print >> oFile, "|".join([str(x) for x in tokens])
oFile.close()
progressMessage("# maq %s\n", i, 28395347)
os.system("""sqlite3 alignedReads.db '.import "tmp/PlatySolexa.txt" PlatySolexa'""")
# dbFilename = ':memory:'
session = initDatabase(dbFilename, iFilename)
# 2a. Parse read alignment file and has results
# 2b. Write out unannotated reads
flankSize = 1000
maqFilename = '/Users/papenfuss/databases/platypus/venom/solexa/mapview.txt'
data = {}
unannFile = open('test_unannotated.txt', 'w')
multFile = open('test_multiple.txt', 'w')
multFile2 = open('test_multiple_gene.txt', 'w')
for i,m in enumerate(MaqViewFile(maqFilename)):
if i==10000: break
if (i % 1000)==0:
progressMessage('# maq alns %s', i)
q = session.query(Gene).filter(Gene.chrom==m.chrom) \
.filter(Gene.start<m.start+flankSize) \
.filter(Gene.end>m.start+32-flankSize).all()
if len(q)==0:
print >> unannFile, m
continue
elif len(q)>1:
x = set([r.geneId for r in q])
if len(x)>1:
print >> multFile, m
print >> multFile2, "%s\t%s" % (m.name, ','.join(x))
continue
tileSize = 35
chrSizeFilename = '/Users/papenfuss/databases/chromSizes/ornAna5.txt'
chrSizes = loadChrSizes(chrSizeFilename)
iFile = open(iFilename)
headers = iFile.readline().strip().split('\t')
oFile = open(oFilename, 'w')
format = "%s\t%i\t%i"
chrom = None
lastChrom = None
countDict = {}
for i,line in enumerate(iFile):
if (i % 1000)==0:
progressMessage('# reads %s', i, 28000000)
tokens = line.strip().split('\t')
d = dict(zip(headers, tokens))
chrom = d['chrom']
if chrom=='MT':
continue
elif 'Ultra' in chrom or 'Contig' in chrom:
pass
else:
chrom = 'chr%s' % chrom
start = int(d['start'])
if chrom!=lastChrom and countDict:
print chrom
for _wStart in xrange(1, chrSizes[lastChrom], tileSize):
counts = countDict.get((lastChrom, _wStart), 0)
dbFilename = os.path.join(iDir, 'mart_names_locations.sqlite')
# dbFilename = ':memory:'
session = initDatabase(dbFilename, iFilename)
# 2a. Parse read alignment file and has results
# 2b. Write out unannotated reads
flankSize = 1000
maqFilename = '/Users/papenfuss/databases/platypus/venom/solexa/mapview.txt'
data = {}
unannFile = open('test_unannotated.txt', 'w')
multFile = open('test_multiple.txt', 'w')
multFile2 = open('test_multiple_gene.txt', 'w')
for i, m in enumerate(MaqViewFile(maqFilename)):
if i == 10000: break
if (i % 1000) == 0:
progressMessage('# maq alns %s', i)
q = session.query(Gene).filter(Gene.chrom==m.chrom) \
.filter(Gene.start<m.start+flankSize) \
.filter(Gene.end>m.start+32-flankSize).all()
if len(q) == 0:
print >> unannFile, m
continue
elif len(q) > 1:
x = set([r.geneId for r in q])
if len(x) > 1:
print >> multFile, m
print >> multFile2, "%s\t%s" % (m.name, ','.join(x))
continue
